[EVPN VXLAN] WHen LIF is moving to new VN control node is not sending evpn route
Affects | Status | Importance | Assigned to | Milestone | ||
---|---|---|---|---|---|---|
Juniper Openstack | Status tracked in Trunk | |||||
R4.1 |
Invalid
|
High
|
Nischal Sheth | |||
Trunk |
Invalid
|
High
|
Nischal Sheth |
Bug Description
Description:
WHen LIF is moving to new VN control node is not sending evpn route
Details:
Lif was initially part of vn-test1. When moving the VN to vn-test2, QFX is subscribing for new VN route target. But CN is not sending any evpn route to QFX.
Workaround:
If we clear the BGP neighbor all routes are send my CN correctly
root@5b11-qfx2> show route table bgp.rtarget.0 advertising-
bgp.rtarget.0: 8 destinations, 15 routes (8 active, 0 holddown, 0 hidden)
Prefix Nexthop MED Lclpref AS path
0:0:0/0
* Self 100 I
65000:65000:1/96
* Self 100 I
65000:
* Self 100 I
65000:
*
65000:
* Self 100 I
65000:
* Self 100 I
65000:
* Self 100 I
root@5b11-qfx2# run show route table default-
3:172.16.
3:172.16.
3:172.16.
3:172.16.
Changed in juniperopenstack: | |
importance: | Undecided → Critical |
importance: | Critical → High |
assignee: | nobody → Nischal Sheth (nsheth) |
milestone: | none → r4.1.0.0-fcs |
information type: | Proprietary → Public |
As discussed offline, this is due to the 0:0:0/0 prefix in bgp.rtarget.0
advertised by QFX. Will mark the bug as Invalid.